U.S. patent number 10,126,152 [Application Number 15/658,435] was granted by the patent office on 2018-11-13 for fluid flow meter with linearization.
This patent grant is currently assigned to Ecolab USA Inc.. The grantee listed for this patent is Ecolab USA Inc.. Invention is credited to Eugene Tokhtuev.
United States Patent |
10,126,152 |
Tokhtuev |
November 13, 2018 |
**Please see images for:
( Certificate of Correction ) ** |
Fluid flow meter with linearization
Abstract
A fluid flow meter is described, that includes intermeshing
gears that may rotate synchronously. The fluid flow meter may
produce a pulsed output that can be normalized to suitable values
according to a method of normalizing input pulses generated in
response to the rotation of gears. A volume counter can be
incremented by an amount equal to a volume per input pulse each
time an input pulse is generated. When the volume counter exceeds a
first reference volume, a normalized output pulse can be generated
until the volume counter exceeds a second reference volume.
Inventors: |
Tokhtuev; Eugene (Duluth,
MN) |
Applicant: |
Name |
City |
State |
Country |
Type |
Ecolab USA Inc. |
St. Paul |
MN |
US |
|
|
Assignee: |
Ecolab USA Inc. (St. Paul,
MN)
|
Family
ID: |
63165512 |
Appl.
No.: |
15/658,435 |
Filed: |
July 25, 2017 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01F
25/0007 (20130101); G01F 1/06 (20130101); G01F
3/10 (20130101); G01F 1/66 (20130101); G01F
25/003 (20130101) |
Current International
Class: |
G01F
1/06 (20060101); G01F 25/00 (20060101); G01F
3/10 (20060101); G01F 1/66 (20060101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
2859228 |
|
Feb 2016 |
|
CA |
|
202188872 |
|
Apr 2012 |
|
CN |
|
202734883 |
|
Feb 2013 |
|
CN |
|
102008008427 |
|
Nov 2009 |
|
DE |
|
2793977 |
|
Nov 2015 |
|
EP |
|
1384789 |
|
Feb 1975 |
|
GB |
|
2120792 |
|
Dec 1983 |
|
GB |
|
2177802 |
|
Jan 1987 |
|
GB |
|
2012126473 |
|
Sep 2012 |
|
WO |
|
2014144557 |
|
Sep 2014 |
|
WO |
|
Other References
Eugene Tokhtuev, U.S. Appl. No. 15/658,437, entitled "Fluid Flow
Meter With Normalized Output," filed Jul. 25, 2017, 34 pages. cited
by applicant .
Hejin Liu et al., Machine translation of the Description, Technical
Field, Summary, Brief Description and Detailed Description of
Chinese Patent Application No. 202188872, filed Apr. 11, 2012, 4
pages. cited by applicant .
Josiah Firth et al., "A novel optical telemetry system applied to
flowmeter networks," Flow Measurement and Instrumentation 48 (2016)
pp. 15-19, Sydney, Australia. cited by applicant .
Suzanne Shelley, "Choosing the Best Flowmeter: Here are the pros
and cons of six popular flowmeter technologies," Chemical
Engineering: New York, 106.7 (Jul. 1999), 13 pages. cited by
applicant.
|
Primary Examiner: Dowtin; Jewel V
Attorney, Agent or Firm: Fredrickson & Byron, P.A.
Claims
The invention claimed is:
1. A fluid flow meter, comprising; a flow chamber; a first gear
intermeshing with a second gear, the first gear and the second gear
being positioned within the flow chamber, the intermeshing of the
first gear and the second gear permitting synchronous rotation of
the first gear and the second gear in response to flow of a fluid
through the flow chamber the fluid flow meter having a nominal
operating range of volumes comprising a maximum volumetric flow
rate and a minimum volumetric flow rate; and a flow sensor
configured to generate a detection signal in response to the
passage of fluid through the flow chamber and/or synchronous
rotation of the first gear and the second gear; and a controller
operatively coupled to a data storage medium, the controller being
configured to: receive detection signal from the flow sensor to
generate input pulses, a volume per input pulse being generally
constant in the nominal operating range, determine an input pulse
frequency, whereby, the input pulse frequency corresponds to a
number of pulses per second, determine whether the fluid flow meter
is operating outside the nominal operating range based on the input
pulse frequency, generate a correction function based on the input
pulse frequency, and apply the correction function to input pulses
when the controller determines that the fluid flow meter is
operated outside the nominal operating range.
2. The fluid flow meter of claim 1, wherein the controller is
configured to store the determined correction function in the data
storage medium.
3. The fluid flow meter of claim 2, wherein the controller is
configured to store the correction function in the data storage
medium in the form of a look-up table.
4. The fluid flow meter of claim 3, wherein the controller is
configured to correlate the correction function to a given pulse
frequency of input pulses, and wherein the data storage medium
stores the correction function corresponding to the given rate of
input pulses.
5. The fluid flow meter of claim 1, wherein the controller is
configured to generate output pulses corresponding to input pulses,
whereby each output pulse is generated by retrieving the correction
function corresponding to the input pulse, and applying the
correction function to a corresponding input pulse.
6. The fluid flow meter of claim 5, wherein the controller is
configured to generate a single output pulse corresponding to a
plurality of input pulses.
7. The fluid flow meter of claim 5, wherein the controller is
configured to delay generation of output pulses when the flow meter
is operating outside the nominal operating range.
8. The fluid flow meter of claim 7, wherein the controller is
configured to delay generation of output pulses by an amount
corresponding to the correction function.
9. A fluid flow meter, comprising; a flow chamber; a first gear
intermeshing with a second gear, the first gear and the second gear
being positioned within the flow chamber, the intermeshing of the
first gear and the second gear permitting synchronous rotation of
the first gear and the second gear in response to flow of a fluid
through the flow chamber the fluid flow meter having a nominal
operating range of volumes comprising a maximum volumetric flow
rate and a minimum volumetric flow rate; and a flow sensor
configured to generate a detection signal in response to the
passage of fluid through the flow chamber and/or synchronous
rotation of the first gear and the second gear; and a controller
operatively coupled to a data storage medium configured to store
generic calibration of the fluid flow meter, wherein the generic
calibration represents a predetermined relationship between a
volumetric flow rate of fluid correlated to volume per input pulse
the controller being configured to: retrieve the generic
calibration of the fluid flow meter, determine a correction
function based on the generic calibration, the correction function
being a function of time interval between input pulses, receive
detection signal from the flow sensor to generate input pulses, and
apply the correction function to the time interval between input
pulses, and generate an output pulse.
10. The fluid flow meter of claim 9, wherein the fluid flow meter
is configured to determine the predetermined time interval between
input pulses based on a slope of the generic calibration.
11. The fluid flow meter of claim 9, wherein the fluid flow meter
is operable in a nominal operating range, wherein a volume of fluid
flowing through the fluid flow meter per actual input pulse is a
generally constant value in the nominal operating range.
12. The fluid flow meter of claim 11, wherein the pulse frequency
represents an inverse of the actual time interval between input
pulses, the input pulse frequency increases monotonically with
respect to the volumetric flow rate in the nominal operating
range.
13. The fluid flow meter of claim 11, wherein the controller is
configured to correct the input pulses when the fluid flow meter
operates outside the nominal operating range.
14. A fluid flow meter, comprising; a flow chamber; a first gear
intermeshing with a second gear, the first gear and the second gear
being positioned within the flow chamber, the intermeshing of the
first gear and the second gear permitting synchronous rotation of
the first gear and the second gear in response to flow of a fluid
through the flow chamber; and a flow sensor configured to generate
a detection signal in response to the passage of fluid through the
flow chamber and/or synchronous rotation of the first gear and the
second gear a controller having an output pulse generator and being
configured to receive detection signal from the flow sensor and
generate input pulses in response to the detection signal,
determine, based on an actual time interval between input pulses
whether the fluid flow meter is operating outside a nominal
operating range, generate a correction function based on a
predetermined time interval between input pulses and the actual
time interval between input pulses, increment a volume counter by
an amount equal to a volume per input pulse at a time corresponding
to time interval of input pulse corrected by the correction
function, transition the output pulse generator from a state where
the output pulse generator does not generate output pulses to a
state where the output pulse generator starts generating an output
pulse when the volume counter exceeds a first reference volume, and
transition the output pulse generator from a state where the output
pulse generator generates the output pulse back to a state where
the output pulse generator stops generating the output pulse.
15. The fluid flow meter of claim 14, wherein the output pulse
generator is configured to generate the output pulse such that the
volume per output pulse is an integer.
16. The fluid flow meter of claim 14, wherein the controller is
configured to generate a single output pulse for every "N" input
pulses generated, whereby N is an integer greater than one.
17. The fluid flow meter of claim 14, wherein the controller is
configured to determine whether volume counter corresponds to a
first reference volume, and if the volume counter corresponds to
the first reference volume, the output pulse generator is further
configured to generate a single output pulse until the volume
counter corresponds to a second reference volume.
18. The fluid flow meter of claim 17, wherein, when the volume
counter exceeds the second reference volume the controller is
further configured to reset the volume counter to zero.
19. The fluid flow meter of claim 14, wherein a pulse frequency of
the output pulse is less than a pulse frequency of input
pulses.
20. The fluid flow meter of claim 14, wherein the output pulses
corresponding to the nominal operating range and output pulses
generated when the flow meter operates outside the nominal
operating range have generally equal pulse characteristics, the
pulse characteristics including at least one of pulse duration, and
duty cycle.
Description
BACKGROUND
Positive displacement fluid measurement systems may be used to
measure a flow rate or volume of a fluid or gas. For example,
dispensing systems may use feedback from a positive displacement
fluid meter to control the volume of fluid dispensed. Such control
systems are often used in lieu of time-on controls to more
accurately dispense precise amounts of fluid or gas and is commonly
used in a variety of settings including, but not limited to, the
industrial, healthcare, pharmaceutical and food and beverage
industries. For example, a positive displacement fluid meter may be
used in the manufacturing process of a drug which requires accurate
measurement of two materials to be mixed into a single batch. The
positive displacement fluid meter may be installed in the supply
lines of the respective materials and feedback from the meters may
be used to dispense the appropriate amount of each material into a
blend tank to be mixed. This application of a positive displacement
meter, like many others, may require the positive displacement
meter to have an accuracy of measurement (e.g., +/-0.5%) to comply
with quality control or regulations, for example. Accordingly, a
positive displacement meter that accurately measures a volume of
fluid or gas can help facilitate performing intended function of a
fluid dispensing system or process.
An example fluid flow meter is described in the commonly-assigned
application, U.S. Pat. No. 9,383,235, assigned to Ecolab Inc., St.
Paul, Minn., the disclosure of which is hereby incorporated by
reference. Manufacturers typically provide a factory calibration
which correlates the volume of a pocket of fluid to a rotational
count corresponding to rotation of one or more components (e.g.,
oval gears) in the flow meter for various volumes of flows. Thus,
by counting the number of pulses produced by the fluid flow meter,
the volume flow rate can be determined based on the factory
calibration.
Such factory calibration may not be accurate outside of a flow
range. For instance, at low flow rates near the flow minimum, the
flow meter may not produce any input pulses, but may still have
flow through various mechanical components of the flow meter.
Similar issues may occur at operation near the flow maximum.
Accordingly, manufacturers specify a range over which the flow
meter's calibration is reliable. However, doing so may be
restrictive to an end user who may intend to use flow meters for
measuring flow over a wide range.
SUMMARY
In one aspect, this disclosure includes a fluid flow meter
comprising a first gear intermeshing with a second gear and thereby
having synchronous rotation in response to flow of a fluid
therethrough. The fluid flow meter having a nominal operating range
between a maximum volumetric flow rate and a minimum volumetric
flow rate. The meter can have a flow sensor to generate a detection
signal in response to the passage of fluid through the flow chamber
and/or synchronous rotation of the first gear and the second gear.
The meter can also have a controller operatively coupled to a data
storage medium. The controller can receive detection signal from
the flow sensor to generate input pulses and determine a pulse
frequency of input pulses, whereby, the pulse frequency corresponds
to a number of pulses per second. The controller can further
determine a deviation of the pulse frequency of input pulses from a
predetermined pulse frequency. The controller can generate a
correction function based on the deviation. The controller can
determine whether the fluid flow meter is operating outside the
nominal operating range based on the deviation and/or the
correction function and apply the correction function to input
pulses when the controller determines that the fluid flow meter is
operated outside the nominal operating range.
In another aspect, the controller can retrieve the generic
calibration of the fluid flow meter. The controller may further
determine a predetermined time interval between input pulses based
on the generic calibration. In addition, the controller can receive
detection signal from the flow sensor to generate actual input
pulses. The controller can determine an actual time interval
between input pulses during use and determine a deviation between
the predetermined time interval and actual time interval between
input pulses. The controller may correct the input pulses by an
amount corresponding to the deviation to generate an output
pulse.
In another aspect, the controller may be coupled to an output pulse
generator. The controller can generate a correction function based
on a predetermined time interval between input pulses and the
actual time interval between input pulses. The controller can
increment a volume counter by an amount equal to a volume per input
pulse at a time corresponding to time interval of input pulse
corrected by the correction function. The controller can transition
the output pulse generator from a state where the output pulse
generator does not generate output pulses to a state where the
output pulse generator starts generating an output pulse when the
volume counter exceeds a first reference volume. The controller can
transition the output pulse generator from a state where the output
pulse generator generates the output pulse back to a state where
the output pulse generator stops generating the output pulse.
The details of one or more examples are set forth in the
accompanying drawings and the description below. Other features,
objects, and advantages will be apparent from the description and
drawings, and from the claims.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a schematic of a fluid flow meter according to an
embodiment;
FIG. 2 is a cross-sectional side view of the fluid flow meter taken
along the sectional plane A-A illustrated in FIG. 1;
FIG. 3A is a sectional-plan view illustrating fluid flow through
the fluid flow meter at a first rotational position of the oval
gears;
FIG. 3B is a sectional-plan view illustrating fluid flow through
the fluid flow meter at a second rotational position of the oval
gears;
FIG. 4A is another sectional-plan view illustrating the fluid flow
meter with non-contact sensors;
FIG. 4B is a schematic illustrating detection signals generated by
the non-contact sensors shown in FIG. 4A;
FIG. 4C is a schematic illustrating a pulse-generation method
according to an embodiment;
FIG. 4D is a schematic illustrating a sequence of valid rotational
states of the oval gears of the fluid flow meter according to a
non-limiting illustrative embodiment;
FIG. 5 is a non-limiting example of a generic calibration
illustrating the relationship between volume per pulse and
volumetric flow rate;
FIG. 6 is an exemplary algorithm to correct generic non-linearities
according to a non-limiting illustrative embodiment;
FIG. 7 is a non-limiting exemplary graph illustrating the
relationship between time between input pulses and volumetric flow
rate;
FIG. 8 is a non-limiting exemplary graph illustrating the
relationship between input pulse frequency and volumetric flow
rate;
FIG. 9 is a non-limiting exemplary graph illustrating the
relationship between volume per pulse and time between input
pulses; and
FIG. 10 is a non-limiting exemplary graph illustrating a function
to correct input pulses.
DETAILED DESCRIPTION
FIG. 1 is a top plan view of a fluid flow measurement system 10
including a fluid flow meter 100. System 10 includes a fluid pump
12, a first fluid line 14, a second fluid line 16 and a fluid flow
meter 100. First fluid line 14 may be in fluid communication with
fluid pump 12 configured to provide a fluid flow through system 10.
Fluid pump 12 may be in fluid communication with a fluid source
(not shown) and may be any suitable pump to provide a fluid flow
through the system. The fluid flow may have a variety of fluid flow
characteristics and may depend on the type of pump selected or the
application of system 10. For example, different applications may
require either a high fluid volume or a low fluid volume. Certain
examples may require uniform fluid flow provided by a peristaltic
pump or pressure-maintained fluid lines. In other examples, a fluid
pump 12 may provide non-uniform fluid flow particularly where the
application requires a low fluid volume.
Fluid flow meter 100 may be configured to measure fluid flow
through system 10 and may include a housing 102 defining a chamber
106, a fluid inlet 104 and a fluid outlet 105. In the illustrated
embodiment, fluid flow meter 100 is a positive displacement meter,
such as an oval gear 108 flow meter. Fluid inlet 104 may be in
fluid communication with first fluid line 14 and provides fluid
flow from the first fluid line 14 into chamber 106. Oval gears 108
and 110 are installed within chamber 106 and are configured to
rotate in concert about fixed axes of rotation 112 and 114,
respectively, in response to fluid flow through the chamber 106.
Fluid exits chamber 106 by way of fluid outlet 105 which is in
fluid communication with second fluid line 16.
Accordingly, fluid provided by fluid pump 12 flows through fluid
line 14 and into fluid flow meter 100 through fluid inlet 104. The
fluid then flows through fluid flow meter 100, wherein the volume
is measured, and out of the fluid flow meter 100 through fluid
outlet 105 and into second fluid line 16.
FIG. 2 is a cross-sectional side view of the fluid flow meter 100
taken along line A-A shown in FIG. 1. Oval gears 108 and 110
installed within the chamber 106 defined by housing 102 and may be
configured to rotate about axes 113 and 115, respectively. In the
illustrated embodiments, fluid flow meter 100 may include flow
sensor 140 and controller 141. The flow sensor 140 may be in
communication (e.g., electrically by way of connection 143, or
wirelessly) with the controller 141. Flow sensor 140 may be
configured to sense a detectable area 146 (not shown) provided on
top surfaces 142 and 144 of oval gears 108 and 110, respectively.
For example, flow sensor 140 may be a magnetic sensor configured to
sense a detectable area 146 comprising a magnet installed on or
within at least one of the oval gears 108. In another example, flow
sensor 140 may be an optical sensor configured to emit a wavelength
onto at least one top surface 142 or 244 of the oval gears 108
including a detectable area 146 and sense a reflectance of the
wavelength off at least one of the top surfaces 142. U.S. Pat. No.
7,523,660, filed Dec. 19, 2007, and U.S. Pat. No. 8,069,719, filed
Feb. 11, 2009, provide examples of oval gears 108 incorporating
non-contact sensors, the entire disclosure of each of which is
hereby incorporated herein by reference. It can be appreciated that
fluid flow meter 100 may include any number of non-contact sensors
and any number of detectable areas suitable for a particular
application of the meter. Flow sensor 140 may also be configured to
generate a detection signal based on the detection, or lack of
detection, of a detectable area 146.
Fluid flow meter 100 may also include controller 141 configured to
calculate a volume of fluid flow through the meter based on the
detection signal of flow sensor 140. The controller 141 may be
configured to receive a detection signal of flow sensor 140 and
generate input pulses to correspond to the rotation of the oval
gears 108 based on the detection signal. The controller 141 can be
a programmable computer such as a microprocessor, a programmable
logic controller 141, and the like, and can include (and/or be in
communication with) non-transitory storage media (e.g., memory or a
non-transitory storage medium 150) for storing instructions in the
form of algorithms and/or data (e.g., calibration data). While an
electrical connection 151 between the controller 141 and a
non-transitory storage medium 150 is illustrated, it should be
understood that the wireless connections between the controller 141
and the non-transitory storage medium 150 are contemplated.
Further, it should be understood that while the electrical
connections of the controller 141, non-transitory storage medium
150 and the fluid flow meter 100 are illustrated as being outside
the housing 102 of the fluid flow meter 100 in FIG. 1, in FIG. 2,
the controller 141 and the non-transitory storage medium 150 (along
with associated connections) are housed within the housing 102 of
the fluid flow meter 100 (as shown in FIG. 2). As will be discussed
further herein, a volume of fluid passing through the fluid flow
meter 100 may be calculated when the number of rotations (complete
and partially complete) made by the oval gears 108 is known and a
volume of fluid per rotation is known. Accordingly, controller 141
may be able to measure a volume of fluid passing through the meter
based on the input pulses generated by the controller 141. In such
cases, controller 141 may include a non-transitory storage medium
150 that stores a calibration between input pulses generated and
volume of fluid passing through the fluid flow meter 100.
FIGS. 3A and 3B are sectional-plan views illustrating fluid flow
through the fluid flow meter 100. As seen therein, oval gears 108
and 110 are configured to intermesh thereby reducing the chances of
fluid from fluid inlet 104 to pass between the gears. Accordingly
fluid flows around the oval gears 108 by way of fluid pockets 116
and 118. FIG. 3A shows fluid flow meter 100 in a first rotational
position where in fluid may be introduced into chamber 106 through
fluid inlet 104. As noted above, the intermeshing of oval gears 108
and 110 reducing the chances of fluid from passing in between the
gears thereby forcing the incoming fluid towards oval gear 108 and
urging oval gear 108 to rotate in a counter-clockwise direction.
The counter-clockwise torque applied across oval gear 108 in turn
urges the clockwise rotation of oval gear 110.
FIG. 3B shows fluid flow meter 100 in a radially advanced
rotational position relative to the rotational position shown in
FIG. 3A, wherein oval gear 108 has rotated 90 degrees
counter-clockwise and oval gear 110 has rotated 90 degrees
clockwise. In this rotational position of fluid flow meter 100, the
rotation of oval gear 108 has formed fluid pocket 118 defined by
the surface of oval gear 108 and a wall of chamber 106.
Concurrently, fluid from fluid inlet 104 is forced toward a vertex
111 of oval gear 110 thereby urging oval gear 110 to rotate in a
clockwise direction. This in turn urges oval gear 108 to continue
rotation in a counter-clockwise direction to release the fluid in
fluid pocket 118. It can be appreciated that a similar fluid pocket
116 may be formed between oval gear 110 and a wall of chamber 106,
as shown in FIG. 3A.
Fluid flow meters according to the present embodiments may be
configured to increase the resolution of measurement thereby
allowing a more precise measurement of fluid flow through the
meter. These configurations may be useful in low fluid flow
applications. In one example, the fluid flow meter 100 may be
configured measure half rotations of the oval gears 108 which
correspond to a volume equal to the volume of two fluid pockets
116. In another example, the fluid flow meter 100 may be configured
to measure quarter rotations of the oval gears 108 which correspond
to a volume equal to one fluid pocket 116. The resolution of
measurement of fluid flow meter 100 may also depend on the volume
of fluid pockets 116 of the meter. Generally, fluid pockets 116
with a smaller volume may increase the measurement resolution of an
oval gear 108 as smaller volumes of fluid are dispensed per
rotation of the oval gears 108. Conversely, larger fluid pockets
116 may decrease the resolution as larger volumes of fluid are
dispensed per rotation. It can be appreciated that different
applications may require a different measurement resolution and
examples of the present application may be configured to have a
wide range of resolutions.
FIG. 4A is a sectional plan view of the fluid flow meter 100
including a flow sensor 140 and a detectable area 146. Flow sensor
140 may be configured to sense detectable area 146 provided on a
surface of oval gear 110 and generate a detection signal. Flow
sensor 140 may be mounted in a housing (102, not shown in FIG. 4A)
of fluid flow meter 100 positioned above the top surfaces 142, 144
of oval gears 108 and 110. As indicated in FIG. 4A oval gear 108
and 110 are configured to rotate counter-clockwise and clockwise,
respectively, in response to fluid flow through chamber 106. The
rotation of oval gear 110 causes detectable area 146 to pass
through a sensing region of flow sensor 140 that may be located
underneath the sensor. Upon sensing detectable area 146, flow
sensor 140 may generate a detection signal. Thus, a detection
signal of flow sensor 140 may be indicative of a rotational
position of oval gears 108 and 110 wherein detectable area 146 is
underneath flow sensor 140. In this example, flow sensor 140 may be
configured to generate a "positive" signal (e.g., "1" or "high")
when the sensor senses the detectable area 146 and a "negative"
signal (e.g., "0" or "low") when the sensor does not sense the
detectable area 146. It can be appreciated that the detection
signal generated by a flow sensor 140 may be of any form in any
format suitable for indicating a sensing of a detectable area 146.
In certain examples, a flow sensor 140 may be configured to not
generate a detection signal when a detectable area 146 is not
sensed. In such an example, the lack of a signal may still be
indicative of a rotational position wherein the detectable area 146
is not within a sensing region of the sensor. As described
previously, the fluid flow meter 100 may include a controller 141
configured to generate a pulsed output based on the detection
signal provided by flow sensor 140. In this example, fluid flow
meter 100 is configured such that rotation of oval gears 108 and
110 may cause flow sensor 140 to sense detectable area 146. Thus,
the controller 141 may be configured to generate a pulse in
response to the detectable area 146 being sensed by the flow sensor
140, as will be described further below.
FIG. 4B is a plot 190 of a detection signal of flow sensor 140 of
fluid flow meter 100 over time according to an example. More
specifically, plot 190 shows the detection signal of flow sensor
140 sensing detectable area 146 as oval gears 108 and 110 rotate in
a forward direction in response to fluid flow through the meter.
Plot 190 includes time points 191 a, 491 b, 492 a and 492 b.
Initially, the detection signal of flow sensor 140 is low
indicating that oval gears 108 and 110 are in a rotational position
wherein the detectable area 146 is not within a sensing region of
the sensor. The detection signal is high between time points 191 a
and 192 a, and also 191 b and 192 b, and is indicative of
rotational positions of the oval gears 108 wherein the detectable
area 146 is sensed by flow sensor 140. The detection signal becomes
low again between time points 192 a and 191 b, and also after time
point 192 b, and is indicative of rotational positions of the oval
gears 108 wherein the detectable area 146 is not sensed by the
sensor. Additional or fewer rotational positions and/or detectable
areas are contemplated within the scope of the present disclosure
(and as will be described further below).
In this example, the rotational positions of the oval gears 108 in
one full rotation of fluid flow meter 100 may be categorized into
rotation states A and B. Rotation state A comprises all the
rotational positions wherein detectable area 146 is not sensed by
flow sensor 140 and is shown in plot 190 before time point 191 a,
between time points 192 a and 191 b, and also after time point 192
b. Rotation state B comprises all the rotational positions wherein
the detectable area 146 is sensed by the flow sensor 140 and is
shown in plot 190 between time points 191 a and 192 a, as well as
191 b and 192 b. When flow sensor 140 senses rotation state A and
B, it generates a negative and positive detection signal,
respectively. In such examples, the fluid flow meter 100 may
include a controller 141 configured to calculate a volume of fluid
flow through the meter by based on the detection signals provided
by flow sensor 140. As oval gears 108 and 110 rotate in a forward
direction in response to fluid flow through the meter, the gears
eventually reach a rotational position wherein detectable area 146
is within a sensing region of the flow sensor 140. Accordingly,
flow sensor 140 may sense rotation state B. It can be appreciated
as the oval gears 108 continue to rotate in fluid flow meter 100,
flow sensor 140 senses a sequence of rotation states comprising
rotation state A and B, in order. As noted above, flow sensor 140
may be configured to generate a negative detection signal and a
positive detection signal when rotation state A and B are sensed,
respectively, and provide the signals to controller 141.
Concurrently, controller 141 of fluid flow meter 100 is configured
to receive the detection signal from flow sensor 140 and produce a
pulsed output. Upon receiving a detection signal indicative of both
a rotation state and a rotational position of oval gears 108 and
110, the controller 141 determines whether the detection signal is
positive. If the detection signal is positive then the controller
141 may generate one or more input pulses. If the detection signal
is negative, then the controller 141 may not generate any input
pulses. Referring back to FIG. 4B, it can be appreciated that input
pulses can be generated by controller 141 at time points 191 a and
191 b when the detection signal goes from low to high.
Alternatively, the controller 141 can be configured to generate
input pulses when the detection signal goes from high to low (e.g.,
at time points 192 a and 192 b) by modifying step 187 to check to
see if the detection signal is negative.
Embodiments described in FIGS. 1-4B can use algorithms that produce
a pulsed output in response to the rotation of the oval gear 108
flow meter. For instance, in the embodiments described in FIGS.
1-4B, the controller 141 can be programmed with instructions that
cause the controller 141 to generate a pulse. In such cases, the
accuracy and resolution of the flow meter can be improved by
generating input pulses that correspond to individual transition of
the gears from a valid rotational state to another valid rotational
state. FIG. 4C is a flow chart corresponding to one such algorithm
400.
In the example illustrated in FIG. 4C, the oval gear 108 meter can
have eight rotational states for every full rotation of the oval
gears 108. For instance, the eight rotational states can be
referred to as states A, B, C, D, E, F, G and H. FIG. 4D
illustrates a chart that shows valid states in the sequence. In
such cases, the controller 141 of the oval gear 108 meter can be
programmed according to the algorithm of FIG. 4C, whereby the
controller 141 is configured at step 402 to determine if the
rotational state detected (e.g., by the flow sensor 140) is a valid
rotational state. The controller 141 is then configured to
determine (at step 404) if the oval gears 108 transition from a
valid rotational state to another valid rotational state, according
to the chart 4D. If for instance, the oval gears 108 transition
from state A to state B, the controller 141 is configured to
determine that the transition is valid, and generate a pulse at
step 406. If on the other hand, the controller 141 determines that
the transition is invalid (for instance, a state other than the
states listed in right column of FIG. 4D for each corresponding
state), then the controller 141 may not generate a pulse
(corresponding to an error condition 408). Accordingly, in this
example, the controller 141 will be configured to generate eight
input pulses for a full rotation of the gears, corresponding to
eight valid transitions between rotational states. While eight
valid rotational states are illustrated, it should be noted that
additional or fewer rotational states (corresponding to additional
or fewer transitions and input pulses) respectively are
contemplated within the scope of the present disclosure. Such
embodiments facilitate accuracy of measurement and eliminate errors
in measurement due to flow non-uniformities (such as jitter or
backflow).
In certain embodiments, the controller 141 is configured to
generate input pulses of duration shorter than time for transition
from a valid rotational state to the next valid rotational state.
In such cases, if the gears rotate "n" rotations per second, with
"m" valid rotational states, the time taken by the oval gears 108
to transition from a valid rotational state to the next valid
rotational state is given by:
.times. ##EQU00001## In such cases, the controller 141 can be
configured to set generate pulses having an input pulse duration
(T.sub.pulse) less than the transition time from a valid rotational
state to the next valid rotational state:
T.sub.pulse<T.sub.transition Such embodiments may facilitate in
an accurate input pulse count by reducing any overlap that may
occur between transition of gears into one or more rotation states
and input pulse generation. In operation, each time the oval gears
108 transition from a valid rotational state to another valid
rotational state, the controller 141 generates an input pulse
having input pulse duration T.sub.pulse. The time interval between
adjacent pulses can be T.sub.s. In such cases, an input pulse
frequency F.sub.s can be defined, whereby the input pulse frequency
is the inverse of the time interval of adjacent input pulses:
##EQU00002## In the illustrated example, the fluid flow meter 100
has eight valid rotational states (as illustrated in FIG. 4D). If,
for instance, the oval gears 108 have 100 rotation per second, the
transition time from one rotational state to next about 1.25
milliseconds in accordance with the above-expression. Accordingly,
the controller 141 may generate input pulses having a duration of
between about 0.1 and about 0.5 ms. More generally, the input pulse
duration (T.sub.pulse) can be between about 5% and about 50% of the
transition time (T.sub.transition).
Referring again to FIG. 4C, the controller 141 may check, at step
410 if previous input pulses were generated. In such cases, the
controller 141 may determine, at step 412, the time interval
between adjacent input pulses, T.sub.s and frequency F.sub.s at
step 414. At step 418, relevant data, such as input pulse count,
time interval between input pulses and/or the frequency.
While the examples below relate to the illustrated fluid flow meter
100 of FIGS. 1-4B, it should be understood the examples described
herein would apply to other types of positive displacement meters
that produce a pulsed output. In some such example embodiments, the
controller 141 can generate input pulses in response to the passage
of the quantity of fluid through the flow chamber 106, and/or the
displacement of rotating components of the fluid flow meter 100.
For instance, the controller 141 can generate input pulses in
response to the synchronous rotation of the first gear and the
second gear as detected by the non-contact sensors. The controller
141 can also be configured to determine the input pulse frequency
F.sub.s for a wide range of known operating conditions and
volumetric flow rates to establish calibration data that can be
stored in the non-transitory storage medium 150.
In certain embodiments, the fluid flow meter 100 can be calibrated
by supplying a known quantity of fluid therethrough, and
determining the volume per input pulse (also referred to as pulse
rate, "P", e.g., in milliliters/pulse) for a known volumetric flow
rate of fluid. Such methods can be referred to herein as "factory
calibration." For instance, referring now to FIG. 5, an example
calibration graph is illustrated that shows the relation between
the volume per input pulse "P" and volumetric flow rate "V" for an
example fluid flow meter 100. In this example, when volume of fluid
passing through the fluid flow meter 100 may not be sufficient to
cause gear rotation, input pulses may not be produced. At some
non-zero value of volumetric flow rate, fluid begins to flow
through the fluid flow meter 100, and input pulses are generated by
the controller 141 at a non-zero input pulse frequency F.sub.s.
In some such embodiments, flow meters typically have a flow range
(e.g., between a maximum volumetric flow rate V.sub.max and minimum
volumetric flow rate V.sub.min) over which the relationship between
volume flow per input pulse and flow (or flow rate) is linear. At
low flow rates (e.g., less than V.sub.min) the flow meter may not
produce any input pulses as a result of the gears being
non-rotational, but may still have fluid flowing through various
mechanical components of the flow meter. Similar issues may occur
at operation near the flow maximum where the flow meter may not
produce any pulses as a result of slippage, even though a non-zero
volume of fluid flows therethrough. As a result, volume per pulse
"P" departs from its value in the range between maximum volumetric
flow rate V.sub.max and minimum volumetric flow rate V.sub.min.
Accordingly, manufacturers typically specify a "nominal operational
range" 500 of the fluid flow meter 100 over which calibration data
of fluid flow meter 100 is believed to be reliable.
In certain illustrative examples, the above flow behavior may be
generic to several flow meters and hence may be referred to as a
`generic calibration.` FIG. 5 represents one such example of a
generic calibration. In FIG. 5, the calibration can be expressed as
a volume that passes through the flow chamber 106 per input pulse
of the flow meter ("P") plotted against volumetric flow rate ("V").
In some such cases, as illustrated, the volume per input pulse "P"
(referred to as "pulse rate") may have an acceptable deviation. For
example, the pulse rate can have a deviation of .+-.3% over the
nominal operational range (between V.sub.min and V.sub.max). The
generic calibration can be stored in the data storage device (e.g.,
in the form of a look-up table). During use, each time an input
pulse is generated, the controller 141 of the flow meter can
retrieve the calibration data and provide an output in the form of
volume or other related quantities (such as volumetric rate, in
milliliters per second and the like). Outside of the nominal
operational range, as seen in FIG. 5, the volume per input pulse
"P" may have unacceptable values of deviation. With an increase in
volume (e.g., when volume is greater than the volume at maximum
volumetric flow rate V.sub.max), the pulse rate can have large
deviation relative to the pulse rate in the nominal operating range
due to slippage. Further, the volume per input pulse may increase
when the volume is less than minimum volumetric flow rate V.sub.min
as a result of input pulses not being generated at low volumes.
Such effects may be referred to as generic "non-linearities" and
may lead to restricted use of flow meters.
The flow characteristics described above may be generic to several
flow meters, and therefore may be a part of the "generic"
calibration data. Accordingly, curve C.sub.1 in FIG. 5 may be
referred to as a "generic calibration curve." While the flow
characteristics of each individual fluid flow meter 100 may depart
from the generic calibration curve due to manufacturing tolerances,
magnitude of generic non-linearities may be greater than any such
variation between flow meters due to manufacturing tolerances. This
may lead to flow meters being restricted to being operated within
their nominal operational range which may inconvenience users.
Accordingly, in some such non-limiting exemplary embodiments, a
fluid flow meter 100 is provided that can extend the range of
operation by correcting output pulses to account for generic
non-linearities. The correction may be performed by the controller
141 in accordance with methods disclosed herein. FIG. 6 represents
one such illustrative algorithm 600 that would be executed by the
controller 141. The algorithm can be stored, for instance, in a
memory or a data storage medium and can be in the form of
machine-readable and/or executable program or steps.
As seen from FIG. 6, at step 602, the controller 141 can receive
detection signal from the flow sensor 140 to generate input pulses.
At step 604, the controller 141 can determine a pulse frequency
F.sub.s of input pulses, whereby, the pulse frequency corresponds
to a number of pulses per second. At step 604, the controller 141
can also determine a deviation of the pulse frequency F.sub.s of
input pulses from a predetermined pulse frequency F.sub.T. The
predetermined pulse frequency F.sub.T may either be retrieved from
the memory or other data storage medium. Alternatively, the
controller 141 may determine the predetermined pulse frequency
F.sub.T based on the generic calibration, as will be described
below. Advantageously, the controller 141 may, at step 608,
determine whether the fluid flow meter 100 is operating outside the
nominal operating range 500 based on the deviation of time interval
between input pulses T.sub.s or input pulse frequency F.sub.s from
a predetermined time interval T.sub.T or predetermined input pulse
frequency F.sub.T respectively. Further, the controller 141 may
determine whether it is necessary to linearize output at step 610.
At step 612, the controller 141 may generate a correction function
based on time interval between input pulses or the input pulse
frequency. The controller 141 can, at step 614, apply the
correction function to input pulses. At step 616, the controller
141 generates output pulses based on the correction function
applied to input pulses. The output pulses may have desired pulse
characteristics (e.g., pulse rate, input pulse frequency, duty
cycle, and the like) and accounts for generic non-linearities.
In certain examples, the controller 141 may only apply the
correction function when the controller 141 determines that the
fluid flow meter 100 is operated outside the nominal operating
range 500 (seen in FIG. 5). This may result in the controller 141
and in turn the flow meter being more optimized and/or using fewer
computational resources, than if the controller 141 were to correct
each individual pulse within the nominal operating range 500 and
outside the nominal operating range 500. Alternatively, if a higher
accuracy is desired, the controller 141 may correct each individual
pulse within the nominal operating range 500 and outside the
nominal operating range 500.
The method 600 may also include additional steps such as storing
the determined correction function in the data storage medium. In
such cases, the controller 141 may determine and correlate the
correction function corresponding to different flow characteristics
(e.g., pulse rate P, time between input pulses T.sub.s and/or
volumetric flow rate V) and store them in the data storage medium
in the form of a look-up table, so that subsequent uses of the flow
meter may involve simply retrieving the corresponding value of the
correction function when one or more flow characteristics are
known. For instance, in an example, the controller 141 may generate
output pulses corresponding to input pulses, whereby each output
pulse is generated by retrieving the correction function
corresponding to the input pulse frequency F.sub.s, and applying
the correction function to a corresponding input pulse.
As mentioned above, the output pulses may have desired pulse
characteristics. In an example, the controller 141 is configured to
generate a single output pulse corresponding to a plurality of
input pulses (more than one input pulse per output pulse).
Accordingly, a pulse frequency of the output pulse can be less than
a pulse frequency of input pulses. In another example, the output
pulses may be normalized as described in U.S. patent application
Ser. No. 15/658,437, filed Jul. 25, 2017, titled "Fluid Flow Meter
with Normalized Output," the entire contents of which is
incorporated by reference. As described therein, the controller 141
may be in communication with an output pulse generator (160, best
seen in FIG. 1) that can generate the output pulse such that the
volume per output pulse is an integer.
In certain examples, the controller 141 can determine and control
the duty cycle of output pulse. In an example, the controller 141
can increment a volume counter each time an input pulse is
generated. In such an example, the controller 141 can determine
whether volume counter corresponds to a first reference volume, and
if the volume counter corresponds to the first reference volume,
the output pulse generator can generate a single output pulse until
the volume counter corresponds to a second reference volume. In
such examples, when the volume counter exceeds the second reference
volume, output pulse may not be generated. Thus, the pulse
duration, and in turn, duty cycle of output pulses may be adjusted
by the controller 141, so as to produce output pulses that are
normalized (e.g., an integer value of volume per output pulse).
As mentioned above, deviations from the nominal operating range 500
may occur when the volumetric flow rate is low, so as to not result
in any input pulses. In such cases, the pulse frequency F.sub.s of
input pulse (number of pulses per second) can be low, and in some
examples, may be nearly zero. As is apparent, when pulse frequency
F.sub.s approaches zero, the time interval between input pulses
T.sub.s which is the inverse of the pulse frequency F.sub.s may
start approaching a large value. By determining time interval
between input pulses T.sub.s or the pulse frequency F.sub.s, the
controller 141 may determine whether the flow meter is operating
outside the nominal operating range 500.
In an example, the controller 141 can generate a correction
function based on input pulse frequency F.sub.s and/or time
interval between input pulses T.sub.s. FIG. 7 illustrates an
exemplary relationship between time interval between input pulses
T.sub.s and volumetric flow rate V and FIG. 8 illustrates an
exemplary relationship between frequency F.sub.s and volumetric
flow rate V. FIGS. 7-8 may be determined from the generic
calibration curve C1 shown in FIG. 5. In an example, the correction
function can be generated based on the relationhip between time
between pulses T.sub.s and volumetric flow rate (shown in FIG. 7),
and pulse rate P and volumetric flow rate (shown in FIG. 5). By
correlating the data between FIGS. 5 and 7, FIG. 9, which
illustrates the relationship between time interval between pulses
and pulse rate can be generated. In this example, the correction
function f corresponds to the graph illustrated in FIG. 9, and can
be used to retrieve a correct pulse rate based on time interval
between pulses, which can then be used to retrieve the correct
volumetric flow rate (from FIG. 5). Thus, flow related quantities
derived from the generic calibration may be correlated to each
other and used for determining the correction function, as
described below. It should be noted that FIGS. 5-9 are non-limiting
illustrative examples and the numerical values and the mathematical
relationship shown by these figures should not be construed as
limiting the scope of the claims of the present application.
For example, referring to FIG. 9, the time interval between input
pulses can be T.sub.s, and the volume per input pulse can be "P".
The volume per input pulse may be known from generic calibration,
and in the nominal operating range 500, may be a generally have a
low value of deviation (e.g., vary by .+-.3% in the nominal
operating range). In some such cases, the volume per input pulse
may be related to time between pulses T.sub.s, as follows:
P=f(T.sub.s)
In the above expression, "f" may refer to a mathematical function.
When multiple input pulses are generated, total volume (e.g., over
"N" pulses) may be represented as follows:
.times. ##EQU00003## Based on the relationship between flow per
input pulse "P", and the time interval between input pulses
T.sub.s, the volume corresponding to several input pulses may be
represented as follows:
.times..function. ##EQU00004## Thus, the controller 141 may be able
to determine volume based on time interval between input pulses.
The controller 141 can use, in this example, the illustrated
relationship between flow per input pulses P and volumetric flow
rate V to determine the correction function "f," as described
further below.
In certain examples, the controller 141 can determine the function
"f" that correlates time interval between input pulses T.sub.s and
volumetric flow rate V based on the generic calibration shown in
FIG. 5. In such cases, the controller 141 can retrieve the generic
calibration of the flow meter (e.g., from memory or data storage
medium). The controller 141 can determine the function "f" from the
generic calibration curve (e.g., seen in FIG. 5) as follows:
f(T.sub.s)=Pulse rate (P)/Volumetric flow rate (V) The correction
function "f", thus correlates the pulse rate P to time between
pulses T.sub.s. The correlation between pulse rate P and time
interval between pulses Ts can be stored (e.g., in memory or data
storage medium) in the form of a look-up table. Thus, for example,
if the controller 141 determines that the time between input pulses
T.sub.s is lower or higher than time between input pulses in the
nominal range, the controller 141 can retrieve the calculated
correction function "f" (e.g., from the look-up table) and
determine the pulse rate corresponding to the time between input
pulses T.sub.s. The resulting value of pulse rate can then be used
to retrieve the volumetric flow rate "V" from the form of look-up
table. In use, when the controller 141 receives a detection signal
from the flow sensor 140 to generate actual input pulses, the
controller 141 can determine whether the flow meter is operating
outside the nominal operating range 500 by comparing an actual time
interval between input pulses, T.sub.s to time interval between
input pulses in the nominal operating range. If the controller 141
determines that the flow meter is operating outside its nominal
operating range 500, the controller 141 may retrieve the correct
pulse rate based on the function "f". The corrected input pulses
can be then be used to generate an output pulse, as will be
described below. The exemplary relationship between the time
interval between input pulses T.sub.s and volumetric flow rate
provided in FIG. 8 should not be construed as limiting. Different
mathematical models may be used to determine the correction
function.
FIG. 10 is an illustrative example of a step function used by the
controller 141 to generate an output pulses. A non-limiting example
of an output pulse generated by the controller 141 is shown in FIG.
10. In FIG. 10, the exemplary step function corresponding to the
nominal operation is shown by the set of steps 1002. The step
function corresponding to slower gear rotation (e.g., than in the
nominal range) is shown by the set of steps 1004, while the step
function corresponding to faster gear rotation (e.g., than in the
nominal range) is shown by the set of steps 1006.
A step function such as that illustrated in FIG. 10 can include
steps corresponding to input pulses. For instance, in the nominal
operating range 500, shown by the set of steps 1002, if a full gear
rotation corresponds to six input pulses, the step function may
have six steps. The step function adjusts the pulse characteristics
of the output pulse (e.g., pulse frequency, duty cycle, etc.) based
on reference volumes as will be described further below.
Optionally, the output pulse generated according to the
non-limiting example of FIG. 10 can be "normalized" such that the
volume per output pulse is an integer value as described in U.S.
patent application Ser. No. 15/658,437, filed Jul. 25, 2017, titled
"Fluid Flow Meter with Normalized Output."
Referring to FIG. 10, each "step" of the step function represents
an incremental volume of fluid flowing through the fluid flow meter
100, and each step extends over a time interval (.DELTA.T.sub.1,
.DELTA.T.sub.2, .DELTA.T.sub.3, etc.). In situations where no
correction is desired or where there are no non-linearities, such
as in the nominal operating range 500, time interval .DELTA.T.sub.1
may be generally equal to the time interval between input pulses,
T.sub.s1. At flow conditions where the gears rotate much slower
than in the nominal flow range, the time interval of steps
.DELTA.T.sub.2 may be adjusted be less than time interval of steps
.DELTA.T.sub.1 in the nominal flow range. At flow conditions where
the gears rotate much faster than in the nominal flow range, the
time interval of steps .DELTA.T.sub.3 may be adjusted be greater
than time interval of steps .DELTA.T.sub.1 in the nominal flow
range.
Once the incremental volume reaches first reference volume V.sub.1
(e.g., at time T.sub.1), the controller 141 begins generation of an
output pulse. The controller 141 can increment a volume counter by
an amount corresponding to volume per input pulse (e.g., pulse rate
"P"), each time an input pulse is generated. During this time,
fluid continues to flow through the flow meter, and the controller
141 continues to increment the volume counter by an amount
corresponding to volume per pulse. When the volume reaches a second
reference volume V.sub.2 (e.g., at time T.sub.2), the output pulse
is stopped, and the volume counter is cleared.
Further, the controller 141 may adjust the step function such that
if time interval between input pulses T.sub.s3 is lower (e.g.,
compared to a predetermined time interval in the nominal operating
range 500), the controller 141 delays incrementing the volume
counter. The delay can be an amount corresponding to a deviation
between predetermined time interval T.sub.T (e.g., obtainable from
generic calibration), and the time interval between input pulses
T.sub.s3 during use. Alternatively, the delay can correspond to the
correction function "f" stored in the form of a look-up table, and
simply applied (e.g., added to time interval between input pulses
T.sub.s) as described above. Thus, the "steps" of the step function
span over a larger time interval .DELTA.T.sub.3 when the time
interval between input pulses T.sub.s3 is less than the
predetermined time interval T.sub.T (e.g., faster gear rotation).
Once the volume counter reaches a first reference volume V.sub.1,
the controller 141 may start generating an output pulse.
Similarly, the controller 141 can adjust the step function duration
in situations where the time interval between input pulses T.sub.s2
is larger (e.g., slower gear rotation at low flow rates), so as to
provide steps that extend over a shorter duration .DELTA.T.sub.2.
Accordingly, the controller 141 may produce more steps per output
pulse, as seen from the set of steps 1004, to account for fewer
pulses. The duration of steps can likewise be determined based on
the deviation of time interval between input pulses T.sub.s from
predetermined time interval T.sub.T (e.g., determinable from the
generic calibration). Appreciably, the steps last over a shorter
duration than in the nominal flow range (e.g.,
.DELTA.T.sub.2<.DELTA.T.sub.1). Advantageously, such embodiments
result in output pulses that are uniform over a wide range of gear
rotations. As seen from FIG. 10, the output pulses corresponding to
the three sets of step functions have similar characteristics.
While exemplary correction functions to correct for non-linearities
are illustrated, such examples should not be construed as limiting.
Further, while examples described above illustrate applying the
correction function when the flow meter is operating outside the
nominal operating range 500, the input pulses can also be corrected
when the flow meter is operating within the nominal operating range
500.
Advantageously, fluid flow meters according to the disclosed
embodiments can permit the output of the oval gear meter to be
corrected to account for generic non-linearities the magnitude of
which may be higher than the measurement uncertainties due to
variability in manufacturing (e.g., tolerances) of individual
meters. A further advantage of embodiments of the present
disclosure is the ability to operate the oval gear meter in the
range outside of the nominal range.
Various examples have been described. These and other examples are
within the scope of the following claims.
* * * * *